claude-ecosystem
Claude Code ecosystem expertise. Modules: CLI tool (setup, slash commands, MCP servers, hooks, plugins, CI/CD), extensibility (agents, skills, output styles creation), CLAUDE.md (project instructions, optimization). Actions: configure, troubleshoot, create, deploy, integrate, optimize Claude Code. Keywords: Claude Code, Anthropic, CLI tool, slash command, MCP server, Agent Skill, hook, plugin, CI/CD, enterprise, CLAUDE.md, agentic coding, agent, skill, output-style, SKILL.md, subagent, Task tool, project instructions, token optimization. Use when: learning Claude Code features, configuring settings, creating skills/agents/hooks, setting up MCP servers, troubleshooting issues, CI/CD integration, initializing or optimizing CLAUDE.md files.
Evaluation results
85%
46%Automated Dependency Security Scanning Agent
Agent creation conventions
Agent directory placement
100%
100%
Name field present
100%
100%
Use when pattern
0%
0%
PROACTIVELY keyword
0%
100%
Example block present
0%
100%
Tools field present
0%
100%
Appropriate model selected
0%
100%
Tool count constraint
0%
100%
Boundary constraints section
100%
100%
Pattern-based body structure
0%
41%
Casual activation language
100%
100%
Without context: $0.2794 · 1m 26s · 18 turns · 24 in / 4,116 out tokens
With context: $0.4563 · 1m 34s · 27 turns · 278 in / 4,501 out tokens
98%
70%Database Migration Guidance Skill
Skill creation conventions
Skill directory placement
0%
100%
allowed-tools field
0%
100%
Structured description
0%
100%
Casual activation language
0%
90%
Expert insight over documentation
58%
100%
Pattern-condition structure
20%
100%
Actionable opening lines
37%
87%
Low reference-table ratio
100%
100%
Line count within limit
100%
100%
When-to-use frontmatter
0%
100%
Without context: $0.2727 · 1m 29s · 17 turns · 23 in / 4,164 out tokens
With context: $0.4494 · 1m 57s · 23 turns · 274 in / 5,022 out tokens
92%
Streamline Project Instructions for Production
CLAUDE.md optimization workflow
Under 400 lines
100%
100%
Prefer under 250 lines
100%
100%
40% reduction achieved
100%
100%
uvicorn command retained
100%
100%
pytest command retained
100%
100%
alembic command retained
100%
100%
black command retained
100%
100%
Git workflow removed
100%
100%
Prerequisites removed
100%
100%
Project structure tree removed
0%
0%
uvicorn not duplicated
100%
100%
Docker section removed
100%
100%
Without context: $0.2323 · 1m 20s · 8 turns · 13 in / 4,397 out tokens
With context: $0.6177 · 2m 21s · 24 turns · 31 in / 6,737 out tokens
- Repository
- majiayu000/claude-skill-registry
- Commit
6213d1a
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.